Patents by Inventor Shantanu R. Gupta

Shantanu R. Gupta has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).

  • Patent number: 5778245
    Abstract: A method and apparatus for dynamically allocating entries of microprocessor resources to particular instructions in an efficient manner to efficiently utilize buffer size and resources. The pipelined and superscalar microprocessor is capable of speculatively executing instructions and also out-of-order processing. Resources within the microprocessor include a store buffer, a load buffer, a reorder buffer and a reservation station. The reorder buffer contains a larger set of physical registers and also contains information related to speculative instructions and the reservation station comprises information related to instructions pending execution. The load buffer is only allocated to load instructions and is valid for an instruction from allocation pipestage to instruction retirement. The store buffer is only allocated to store instructions and is valid for an instruction from allocation to store performance.
    Type: Grant
    Filed: March 1, 1994
    Date of Patent: July 7, 1998
    Assignee: Intel Corporation
    Inventors: David B. Papworth, Andrew F. Glew, Glenn J. Hinton, Robert P. Colwell, Michael A. Fetterman, Shantanu R. Gupta, James S. Griffith
  • Patent number: 5689674
    Abstract: A method and apparatus for binding instructions to dispatch ports in a reservation station includes a counter mechanism and a port identifier. The counter mechanism maintains a count of instructions which are pending dispatch from at least one of the dispatch ports. The port identifier receives an instruction and identifies to which of the dispatch ports the instruction is to be bound, based on the count of instructions maintained by the counter mechanism.
    Type: Grant
    Filed: October 31, 1995
    Date of Patent: November 18, 1997
    Assignee: Intel Corporation
    Inventors: James S. Griffith, Shantanu R. Gupta, Glenn J. Hinton
  • Patent number: 5627984
    Abstract: A two cycle pipelined method and apparatus for allocating a number of vacant entries of a buffer resource and generating a set of enable vectors based thereon for a set of issued instructions. The procedure for determining the vacant entries is spread across two pipestages (clock cycles) of a pipelined superscalar processor. For each pipestage, the system receives information from the previous pipestage as to which entries were eligible for allocation but have not yet received instruction information as well as a set of speculative stall signals. For each pipestage, the reservation station informs the system as to which entries are vacant according to the reservation station's knowledge at that time; this is a preliminary deallocation vector. For each pipestage, the system also receives a list of the instructions for allocation to the reservation station for that cycle.
    Type: Grant
    Filed: March 28, 1996
    Date of Patent: May 6, 1997
    Assignee: Intel Corporation
    Inventors: Shantanu R. Gupta, James S. Griffith
  • Patent number: 5584038
    Abstract: An allocator assigns entries for a circular buffer. The allocator receives requests for storing data in entries of the circular buffer, and generates a head pointer to identify a starting entry in the circular buffer for which circular buffer entries are not allocated. In addition to pointing to an entry in the circular buffer, the head pointer includes a wrap bit. The allocator toggles the wrap bit each time the allocator traverses the linear queue of the circular buffer. A tail pointer is generated, including the wrap bit, to identify an ending entry in the circular buffer for which circular buffer entries are allocated. In response to the request for entries, the allocator sequentially assigns entries for the requests located between the head pointer and the tail pointer. The allocator has application for use in a microprocessor performing out-of-order dispatch anti speculative execution. The allocator is coupled to a reorder buffer, configured as a circular buffer, to permit allocation of entries.
    Type: Grant
    Filed: April 17, 1996
    Date of Patent: December 10, 1996
    Assignee: Intel Corporation
    Inventors: David B. Papworth, Andrew F. Glew, Michael A. Fetterman, Glenn J. Hinton, Robert P. Colwell, Steven J. Griffith, Shantanu R. Gupta, Narayan Hegde
  • Patent number: 5584037
    Abstract: An allocator assigns entries for a circular buffer. The allocator receives requests for storing data in entries of the circular buffer, and generates a head pointer to identify a starting entry in the circular buffer for which circular buffer entries are not allocated. In addition to pointing to an entry in the circular buffer, the head pointer includes a wrap bit. The allocator toggles the wrap bit each time the allocator traverses the linear queue of the circular buffer. A tail pointer is generated, including the wrap bit, to identify an ending entry in the circular buffer for which circular buffer entries are allocated. In response to the request for entries, the allocator sequentially assigns entries for the requests located between the head pointer and the tail pointer. The allocator has application for use in a microprocessor performing out-of-order dispatch and speculative execution. The allocator is coupled to a reorder buffer, configured as a circular buffer, to permit allocation of entries.
    Type: Grant
    Filed: December 13, 1995
    Date of Patent: December 10, 1996
    Assignee: Intel Corporation
    Inventors: David B. Papworth, Andrew F. Glew, Michael A. Fetterman, Glenn J. Hinton, Robert P. Colwell, Steven J. Griffith, Shantanu R. Gupta, Narayan Hegde
  • Patent number: 5560025
    Abstract: A method and apparatus for searching for a pattern of values indicating vacancy within a reservation station. The present invention includes a method and apparatus for search a deallocation vector of an instruction scheduler in order to locate, within one clock cycle, a pattern of the first vacancies within the instruction scheduler for storage of instruction information associated with several microprocessor instructions. The present invention advantageously locates four vacant entries of the deallocation vector which specify the first four vacancies within a reservation station of the instruction scheduler and may be utilized to locate the first four vacant entries as well. The present invention performs the above processing utilizing high speed parallel processing methods so that the entire searching, reporting and updating functions, with regard to the deallocation vector, can be completed within one clock cycle. Two embodiments of the present invention, a static and a dynamic embodiment, are presented.
    Type: Grant
    Filed: March 31, 1993
    Date of Patent: September 24, 1996
    Assignee: Intel Corporation
    Inventors: Shantanu R. Gupta, James S. Griffith
  • Patent number: 5524263
    Abstract: A method and apparatus for handling resource allocation during processor stall conditions. The instruction issue components of a processor are stalled (e.g., the issuance of new instruction is frozen) as a result of various stall conditions. One stall condition (full stall) occurs when an allocated buffer resource becomes full. Another stall condition (partial stall) occurs during register renaming and a given instruction sources a larger register width than the register alias table currently contains within the RAT buffer. This is a partial width data dependency and a partial stall is asserted. The present invention, upon detection of a full stall, does not allocate any buffer entries within the clock cycle that causing the full stall and resource pointers are not advanced and instructions issued during that clock cycle are not allocated. Within the clock cycle of the deassertion of the full stall, the resource buffers are allocated and the resource allocation pointers are updated.
    Type: Grant
    Filed: February 25, 1994
    Date of Patent: June 4, 1996
    Assignee: Intel Corporation
    Inventors: James S. Griffth, Shantanu R. Gupta, Narayan Hegde
  • Patent number: 5490280
    Abstract: A method and apparatus for allocating a number of vacant entries of a buffer resource and generating a set of enable vectors based thereon for a set of issued instructions. A deallocation vector of a reservation station is searched in order to locate, within one clock cycle, the vacancies within the reservation station for storage of instruction information associated with several issued operations. Vacancies are indicated by bits of the deallocation vector. A general static and dynamic approach are disclosed for performing the vacant entry identification at high speed within a single clock cycle. Alternate embodiments are disclosed, based on the general approach, that divide the deallocation vector into separate portions (consecutive bits or interleaved) and process each portion based on the general approaches. Rotating priority reference points within the deallocation vector may be used to vary the starting point for vacancy location.
    Type: Grant
    Filed: June 28, 1994
    Date of Patent: February 6, 1996
    Assignee: Intel Corporation
    Inventors: Shantanu R. Gupta, James S. Griffith, Glenn J. Hinton
  • Patent number: 5453708
    Abstract: A clocking scheme provides for an improved latching of an output from a domino circuit by delaying a precharging of a domino node. The precharging delay is achieved by introducing the delay in the clocking circuitry which activates the precharging of the domino node. No delay is introduced in the data path in order not to delay the evaluation and transmission of the data signal. During one phase of a clocking cycle, the domino node is precharged to a predetermined logic state. Also during this precharge phase, an input latch couples an input data signal to the domino circuit. During the other phase of the clocking cycle, the domino circuit performs a logic operation based on the input signal. Also during this evaluation phase, an output latch latches the logic state of the domino output for transmission from the output latch.
    Type: Grant
    Filed: January 4, 1995
    Date of Patent: September 26, 1995
    Assignee: Intel Corporation
    Inventors: Shantanu R. Gupta, Thomas D. Fletcher